This invention relates in general to wireless telecommunications networks and applications and, in particular, to a method and system of optimizing the signal strength and interference characteristics of the network. More particularly, the invention relates to methods of predicting changes in coverage as changes in the carrier-to-interference (C/I) ratio of cells in the network are made, as well as modifying the power levels of wireless components in the network to reach acceptable tradeoffs between interference and coverage.
Without limiting the scope of the invention, its background is described in connection with a wireless telecommunications network utilizing recordations of disturbance events to predict changes and determine tradeoffs between interference and coverage as power levels are modified, as an example.
Present-day mobile telephony has spurred rapid technological advances in both wireless and wireline communications. The wireless industry, in particular, is a rapidly growing industry, with advances, improvements, and technological breakthroughs occurring on an almost daily basis. Many mobile or wireless telecommunications systems, among them the European GSM-system, have passed through several generations of advancements and development phases. System designers are now concentrating on further improvements to such systems, including system refinements and the introduction of optional subscriber services.
Most wireless telecommunications systems are implemented as cellular telephone networks wherein a group of Base Transceiver Stations (BTSs), or base stations are served by a centrally located switch. The switch is commonly referred to as a Mobile Switching Center (MSC). The base stations are spaced apart from each other by distances of between one-half and twenty kilometers. Each base station is assigned a number of two-way voice and control channels. The voice channels transmit voice signals to and from proximately located mobile stations, and transmit control information to and from these mobile stations, usually for the purpose of establishing a voice communications link.
A typical cellular telephone network also includes a group of interconnected MSCs, which operate in association with a gateway mobile switching center (GMSC) through which the cellular telephone network interconnects with a conventional Public Switched Telephone Network (PSTN). In addition, at least one Home Location Register (HLR) operates within the cellular telephone network. The HLR stores network subscriber information, including the identified location of current mobile stations within the network.
In response to an incoming call placed to a mobile station, the MSC queries the HLR to determine the current location of the station. The HLR xe2x80x9clooks upxe2x80x9d the current location of the mobile station and contacts the currently serving MSC to pre-route the call and retrieve a temporary location directory number, which is utilized to route the call through the telecommunications network for termination of the call to the mobile station. The MSC instructs the base station serving the cell in which the mobile station is located to page the mobile station. Responding to the page, the mobile station requests assignment of a channel, and the network terminates the call through the serving MSC and over the assigned channel.
Calls by mobile subscribers can be affected by interference or radio disturbance events which, in turn, limit the efficiency of the network. As such, it is important to identify those cells within the network, which are sources of and subject to radio disturbance events. Interference itself can be either external or internal to the radio network. The internal interference results from call activities within a network cell site. In this regard, it is appropriate to term the cells as either xe2x80x9coffendingxe2x80x9d or xe2x80x9cdisturbed.xe2x80x9d In this regard, a radio disturbance event typically occurs during a cellular call, either on the downlink (from a base station to a mobile station) or on the uplink (from a mobile station to a base station). The disturbance events include co-channel interference or adjacent channel interference. Similarly, different sources of external interference exist that can create problems in the network.
Various methods exist for determining when a disturbance event has occurred. For example, a comparison of signal strength versus a measurement of speech quality can be employed to determine when a low Carrier-to-Interference (C/I) ratio or high Bit Error Rate (BER) exists. When acceptable signal strength is correlated with degraded speech quality for an extended period of time (usually measured in seconds), that cell can be considered xe2x80x9cdisturbed.xe2x80x9d Failure to identify and analyze sources of such disturbances could result in poor channel quality or the sealing of devices in the network, which means they are unavailable for use in handling calls.
Additionally, methods and systems exist for identifying cells within the wireless telecommunications network that are disturbed. One such method involves the use of downlink interference prediction tools, or prediction methods which use model-based prediction algorithms. Such methods predict where interference may exist within a given network coverage area allowing subsequent use of such predictions for cell and frequency planning, particularly in initial network designs. Frequency planning is used to account for frequency reuse by different cells which creates interference in adjacent or neighboring cells.
The validity of such predictions is dependent on a number of factors, including the accuracy of the propagation model utilized, the resolution of the terrain data, and so forth. Such tools are helpful in identifying the cells that are causing downlink interference, but taken together are often inaccurate because of the dependence on predictions. That is, such prediction tools do not always account for xe2x80x9creal-lifexe2x80x9d sources of interferences in the coverage area as determined through more empirical measurement methods. xe2x80x9cReal-lifexe2x80x9d sources may include natural obstructions and buildings.
Currently, radio network engineers use the downlink interference prediction tools, along with trial and error, and xe2x80x9cdrive-byxe2x80x9d techniques to predict and measure the effects of changes in the radio network. These tools are useful in predicting where interference will effect the cellular system given a specific output power generated by the radio base station. The predictions are accurate only if the propagation model is accurate. The sources of interference can be identified, but again, the accuracy is a function of the propagation model. The drive-by methods, on the other hand, are quite accurate as they are based on clinical measurements, but require an immense amount of resources to implement.
Another method utilized to identify the source of interference in a wireless telecommunications network by the same inventors is disclosed and claimed in the related application by the same inventors. In the related application, the inventors describe a method of underlying sources of uplink/downlink interference by considering call events occurring in the offending cells and the resulting disturbance events in a disturbed cell. These event recordings are run in conjunction with each other for the purpose of correlating and computing the two events as a function of time. The time stamps of the call start attempts (voice channel seizure events) result in Call Event Recordings (CER) for possible adjacent channel/co-channel mobiles.
While prior art techniques are useful in identifying, predicting and measuring the effects of interference in the network, they do not provide guidance to the network engineer as to how to best modify the network to improve its efficiency. In particular, the prior art methods do not suggest how changes in one part of the network can effect the performance in another part. What is needed is a method of predicting changes in the disturbed cell with interference created in other offending (co-channel) cells. A means of performing a tradeoff and deciding whether to improve coverage in the disturbed cell or sacrifice it with lesser interference in a co-channel neighbor would provide numerous advantages.
The present invention provides a method and system for optimizing signal strength and interference characteristics in a wireless telecommunications network. With the present invention, once the network operator has identified the sources of interference, such information may be used in improving performance of the network. That is, a tradeoff may be performed which includes improving coverage in the disturbed cell by sacrificing coverage in an offending cell as a function of signal strength and interference.
Disclosed in one embodiment is a method of optimizing the signal strength and interference characteristics in a wireless telecommunications network including a plurality of cells with at least one of the cells being disturbed (the xe2x80x9cdisturbed cellxe2x80x9d) by events occurring in one or more offending cells. The method comprises the step of computing a distribution of disturbance events in the disturbed cell as a function of disturbances created by the offending cells. The distribution is a probability matrix for different carrier(coverage)-to-interference (C/I) ratio levels. As such, a distribution of disturbance events can be computed by computing the signal strength for calls in progress and calls not in progress (on an idle channel) to utilize in calculating C/I ratio levels. In particular, the C/I combinations for particular points in the network are summed to provide a range of C/I.
The C/I ratio levels are then utilized to create a signal strength plan which defines regions in order to perform a cell assessment, as well as to identify an acceptable threshold for the network. For illustrative purposes, 17 dB is shown to be the acceptable threshold. Therefore, for example, Region I corresponds to low disturbance level and low signal strength (C/I less than 17dB), while Region II corresponds to a high disturbance level and low signal strength (C/I less than 17 dB). In one embodiment, five (5) regions are identified in the C/I distribution.
The method also comprises the step of utilizing the distribution of disturbance events to predict changes in coverage of the offending cells as changes in the carrier-to-interference (C/I) ratio of the disturbed cell are made. This includes calculating the improvement (GCI) in C/I ratio level in the disturbed cell and the effect in C/I ratio levels in the offending cells. The improvement is calculated by increasing the signal strength in the offending cells while decreasing the coverage provided by the disturbed cell. Once the increase and decrease has been performed, respectively, the signal strength and interference are then adjusted so as to avoid a significant decrease in coverage in the offending cells.
The improvement may also be calculated by decreasing the signal strength in the offending cells while increasing the coverage provided by the disturbed cell. Again, adjustments must be performed on the signal strength and interference in order to avoid a significant decrease in service provided to mobile stations in the offending cells.
The method further comprises the step of modifying the power levels of wireless components in the network to reach acceptable tradeoffs between interference and coverage. The tradeoff is performed as a function of the predicted changes in C/I, along with acceptable coverage and interference thresholds for the network. In modifying the power levels, a check is performed to verify whether the improved carrier-to-interference ratio level (GCI) for a higher power setting (Power+Delta) is acceptable. If the GCI is positive and the higher power setting is acceptable, then the recommended mobile station power is set to (Power+Delta). If, however, the GCI for a higher power setting (Power+Delta) is not acceptable, then a check is done to determine if the power for a weaker setting is acceptable. If neither weaker nor higher powers are acceptable, then the current setting is retained.
A technical advantage of the present invention includes more accurate predictions for implementing a tradeoff due to the utilization of empirical measurements based on recorded disturbance events and not predictions. As such, the interference analysis takes into account the behavior of all mobile stations within the network rather than a particular mobile event.
Another technical advantage of the present invention is enabling radio network engineers to optimize the network with the objective of minimizing interference.